valknar/runpod
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor: rename docker-compose.gpu.yaml to compose.yaml

Browse Source 
Simplified compose file naming to follow Docker Compose best practices:
- Renamed docker-compose.gpu.yaml to compose.yaml
- Updated all references in documentation files (README.md, DEPLOYMENT.md, GPU_DEPLOYMENT_LOG.md, RUNPOD_TEMPLATE.md)
- Updated references in scripts (prepare-template.sh)

This change enables simpler command syntax:
- Before: docker compose -f docker-compose.gpu.yaml up -d orchestrator
- After: docker compose up -d orchestrator

Generated with Claude Code (https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
Sebastian Krüger
2025-11-21 14:49:42 +01:00
parent cafa0a1147
commit 0fa69cae28
6 changed files with 28 additions and 28 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
README.md
View File

@@ -23,7 +23,7 @@ If you've already created a RunPod template:
1. Deploy pod from template in RunPod dashboard 1. Deploy pod from template in RunPod dashboard
2. SSH to the pod 2. SSH to the pod
3. Create `.env` file with your credentials 3. Create `.env` file with your credentials
4. Start orchestrator: `docker compose -f docker-compose.gpu.yaml up -d orchestrator` 4. Start orchestrator: `docker compose -f compose.yaml up -d orchestrator`
**See**: [RUNPOD_TEMPLATE.md](RUNPOD_TEMPLATE.md) for template usage instructions. **See**: [RUNPOD_TEMPLATE.md](RUNPOD_TEMPLATE.md) for template usage instructions.
@@ -107,7 +107,7 @@ models:
endpoint: /v1/chat/completions endpoint: /v1/chat/completions
``` ```
Then add the Docker service to `docker-compose.gpu.yaml` and restart the orchestrator. Then add the Docker service to `compose.yaml` and restart the orchestrator.
**See**: [docs/DEPLOYMENT.md](docs/DEPLOYMENT.md#adding-new-models) for complete instructions. **See**: [docs/DEPLOYMENT.md](docs/DEPLOYMENT.md#adding-new-models) for complete instructions.

0
docker-compose.gpu.yaml → compose.yaml
View File

24
docs/DEPLOYMENT.md
View File

@@ -94,7 +94,7 @@ ssh gpu-pivoine
cd /workspace/ai/ cd /workspace/ai/
# Start orchestrator (always running) # Start orchestrator (always running)
docker compose -f docker-compose.gpu.yaml up -d orchestrator docker compose -f compose.yaml up -d orchestrator
# Orchestrator will automatically manage model services as needed # Orchestrator will automatically manage model services as needed
``` ```
@@ -217,7 +217,7 @@ models:
### Step 2: Add Docker Service ### Step 2: Add Docker Service
```yaml ```yaml
# Add to ai/docker-compose.gpu.yaml # Add to ai/compose.yaml
services: services:
vllm-llama: vllm-llama:
build: ./vllm build: ./vllm
@@ -245,7 +245,7 @@ services:
```bash ```bash
ssh gpu-pivoine ssh gpu-pivoine
cd /workspace/ai/ cd /workspace/ai/
docker compose -f docker-compose.gpu.yaml restart orchestrator docker compose -f compose.yaml restart orchestrator
``` ```
**That's it!** The orchestrator automatically detects the new model. **That's it!** The orchestrator automatically detects the new model.
@@ -256,13 +256,13 @@ docker compose -f docker-compose.gpu.yaml restart orchestrator
```bash ```bash
# Start orchestrator # Start orchestrator
docker compose -f docker-compose.gpu.yaml up -d orchestrator docker compose -f compose.yaml up -d orchestrator
# View orchestrator logs # View orchestrator logs
docker logs -f ai_orchestrator docker logs -f ai_orchestrator
# Restart orchestrator # Restart orchestrator
docker compose -f docker-compose.gpu.yaml restart orchestrator docker compose -f compose.yaml restart orchestrator
# Check active model # Check active model
curl http://100.100.108.13:9000/health curl http://100.100.108.13:9000/health
@@ -287,10 +287,10 @@ curl http://100.100.108.13:9000/health | jq '.current_model'
```bash ```bash
# Manually start a specific model (bypassing orchestrator) # Manually start a specific model (bypassing orchestrator)
docker compose -f docker-compose.gpu.yaml --profile text up -d vllm-qwen docker compose -f compose.yaml --profile text up -d vllm-qwen
# Stop a model # Stop a model
docker compose -f docker-compose.gpu.yaml stop vllm-qwen docker compose -f compose.yaml stop vllm-qwen
# View model logs # View model logs
docker logs -f ai_vllm-qwen_1 docker logs -f ai_vllm-qwen_1
@@ -337,10 +337,10 @@ docker logs -f ai_musicgen_1
docker logs ai_orchestrator docker logs ai_orchestrator
# Check if model service exists # Check if model service exists
docker compose -f docker-compose.gpu.yaml config | grep -A 10 "vllm-qwen" docker compose -f compose.yaml config | grep -A 10 "vllm-qwen"
# Manually test model service # Manually test model service
docker compose -f docker-compose.gpu.yaml --profile text up -d vllm-qwen docker compose -f compose.yaml --profile text up -d vllm-qwen
curl http://localhost:8001/health curl http://localhost:8001/health
``` ```
@@ -354,8 +354,8 @@ ls -l /var/run/docker.sock
sudo systemctl restart docker sudo systemctl restart docker
# Rebuild orchestrator # Rebuild orchestrator
docker compose -f docker-compose.gpu.yaml build orchestrator docker compose -f compose.yaml build orchestrator
docker compose -f docker-compose.gpu.yaml up -d orchestrator docker compose -f compose.yaml up -d orchestrator
``` ```
### Model Switching Too Slow ### Model Switching Too Slow
@@ -376,7 +376,7 @@ docker run --rm -it --gpus all \
``` ```
ai/ ai/
├── docker-compose.gpu.yaml # Main orchestration file ├── compose.yaml # Main orchestration file
├── .env.example # Environment template ├── .env.example # Environment template
├── README.md # This file ├── README.md # This file
│ │

4
docs/GPU_DEPLOYMENT_LOG.md
View File

@@ -260,7 +260,7 @@ models:
### Deployment Changes ### Deployment Changes
#### Docker Compose Structure #### Docker Compose Structure
- **File**: `docker-compose.gpu.yaml` - **File**: `compose.yaml`
- **Services**: 4 total (1 orchestrator + 3 models) - **Services**: 4 total (1 orchestrator + 3 models)
- **Profiles**: `text`, `image`, `audio` (orchestrator manages activation) - **Profiles**: `text`, `image`, `audio` (orchestrator manages activation)
- **Restart Policy**: `no` for models (orchestrator controls lifecycle) - **Restart Policy**: `no` for models (orchestrator controls lifecycle)
@@ -319,7 +319,7 @@ ssh gpu-pivoine
cd /workspace/ai/ cd /workspace/ai/
# Start orchestrator (manages everything) # Start orchestrator (manages everything)
docker compose -f docker-compose.gpu.yaml up -d orchestrator docker compose -f compose.yaml up -d orchestrator
# Check status # Check status
curl http://100.100.108.13:9000/health curl http://100.100.108.13:9000/health

14
docs/RUNPOD_TEMPLATE.md
View File

@@ -116,14 +116,14 @@ ls -lh /workspace/musicgen/models/
# Test orchestrator starts # Test orchestrator starts
cd /workspace/ai cd /workspace/ai
docker compose -f docker-compose.gpu.yaml up -d orchestrator docker compose -f compose.yaml up -d orchestrator
docker logs ai_orchestrator docker logs ai_orchestrator
# Test model loading (should be fast since models are cached) # Test model loading (should be fast since models are cached)
curl http://localhost:9000/health curl http://localhost:9000/health
# Stop orchestrator # Stop orchestrator
docker compose -f docker-compose.gpu.yaml down docker compose -f compose.yaml down
``` ```
### Step 4: Clean Up Before Saving ### Step 4: Clean Up Before Saving
@@ -215,7 +215,7 @@ EOF
tailscale up --authkey=<YOUR_TAILSCALE_KEY> tailscale up --authkey=<YOUR_TAILSCALE_KEY>
# Start orchestrator (models already cached, starts in seconds!) # Start orchestrator (models already cached, starts in seconds!)
docker compose -f docker-compose.gpu.yaml up -d orchestrator docker compose -f compose.yaml up -d orchestrator
# Verify # Verify
curl http://localhost:9000/health curl http://localhost:9000/health
@@ -295,12 +295,12 @@ docker compose --profile audio up -d musicgen
```bash ```bash
# Build images one at a time # Build images one at a time
docker compose -f docker-compose.gpu.yaml build orchestrator docker compose -f compose.yaml build orchestrator
docker compose -f docker-compose.gpu.yaml build vllm-qwen docker compose -f compose.yaml build vllm-qwen
docker compose -f docker-compose.gpu.yaml build musicgen docker compose -f compose.yaml build musicgen
# Check build logs for errors # Check build logs for errors
docker compose -f docker-compose.gpu.yaml build --no-cache --progress=plain orchestrator docker compose -f compose.yaml build --no-cache --progress=plain orchestrator
``` ```
### Tailscale Won't Install ### Tailscale Won't Install

10
scripts/prepare-template.sh
View File

@@ -118,15 +118,15 @@ build_docker_images() {
# Build orchestrator # Build orchestrator
log_info "Building orchestrator..." log_info "Building orchestrator..."
docker compose -f docker-compose.gpu.yaml build orchestrator docker compose -f compose.yaml build orchestrator
# Build vLLM # Build vLLM
log_info "Building vLLM..." log_info "Building vLLM..."
docker compose -f docker-compose.gpu.yaml build vllm-qwen docker compose -f compose.yaml build vllm-qwen
# Build MusicGen # Build MusicGen
log_info "Building MusicGen..." log_info "Building MusicGen..."
docker compose -f docker-compose.gpu.yaml build musicgen docker compose -f compose.yaml build musicgen
# Pull Flux image (pre-built) # Pull Flux image (pre-built)
log_info "Pulling Flux.1 image..." log_info "Pulling Flux.1 image..."
@@ -196,7 +196,7 @@ validate_installation() {
# Start orchestrator # Start orchestrator
log_info "Starting orchestrator for validation..." log_info "Starting orchestrator for validation..."
docker compose -f docker-compose.gpu.yaml up -d orchestrator docker compose -f compose.yaml up -d orchestrator
sleep 10 sleep 10
@@ -217,7 +217,7 @@ validate_installation() {
fi fi
# Stop orchestrator # Stop orchestrator
docker compose -f docker-compose.gpu.yaml down docker compose -f compose.yaml down
log_success "Validation passed" log_success "Validation passed"
} }